Get member of __m128 by index?

by

A union is probably the most portable way to do this:

union {
    __m128 v;    // SSE 4 x float vector
    float a[4];  // scalar array of 4 floats
} U;

float vectorGetByIndex(__m128 V, unsigned int i)
{
    U u;

    assert(i <= 3);
    u.v = V;
    return u.a[i];
}

More Related Contents:

  1. C++ error: ‘_mm_sin_ps’ was not declared in this scope
  2. Most efficient way to check if all __m128i components are 0 [using
  3. AVX2 what is the most efficient way to pack left based on a mask?
  4. How to solve the 32-byte-alignment issue for AVX load/store operations?
  5. print a __m128i variable
  6. How to implement atoi using SIMD?
  7. How to efficiently perform double/int64 conversions with SSE/AVX?
  8. SIMD prefix sum on Intel cpu
  9. Loading 8 chars from memory into an __m256 variable as packed single precision floats
  10. What’s the difference between logical SSE intrinsics?
  11. SSE, intrinsics, and alignment
  12. Do I get a performance penalty when mixing SSE integer/float SIMD instructions
  13. Emulating shifts on 32 bytes with AVX
  14. Visual Studio 2017: _mm_load_ps often compiled to movups
  15. How to implement “_mm_storeu_epi64” without aliasing problems?
  16. SSE reduction of float vector
  17. Where can I find an official reference listing the operation of SSE intrinsic functions?
  18. Is `reinterpret_cast`ing between hardware SIMD vector pointer and the corresponding type an undefined behavior?
  19. Why is this program erroneously rejected by three C++ compilers?
  20. Using AVX CPU instructions: Poor performance without “/arch:AVX”
  21. How is a vector’s data aligned?
  22. Standard-layout and tail padding
  23. Unoptimized clang++ code generates unneeded “movl $0, -4(%rbp)” in a trivial main()
  24. Overloaded lambdas in C++ and differences between clang and gcc
  25. Clang doesn’t see basic headers
  26. Conflict between copy constructor and forwarding constructor
  27. Most insanely fast way to convert 9 char digits into an int or unsigned int
  28. Getting Clang to work on windows
  29. Injected class name compiler discrepancy
  30. Why is this SIMD multiplication not faster than non-SIMD multiplication?

Leave a Comment