ENH: [1/7] enable multi-platform SIMD compiler optimizations

  Implement new distutils class `CCompilerOpt`, used for handling
  the CPU/hardware optimization, starting from parsing the
  command arguments, to managing the relationship between the CPU baseline
  and dispatch-able features, also generating the required C headers
  and ending with compiling the sources with proper compiler's flags.

  `CCompilerOpt` mainly used as a helper class for `CCompiler`,
  and doesn't provide any runtime detection for the CPU features,
  instead only focuses on the compiler side, but it generates
  abstract C headers that can be used later for the final
  runtime dispatching process.

1 files changed, 8 insertions, 0 deletions

diff --git a/numpy/distutils/checks/cpu_fma3.c b/numpy/distutils/checks/cpu_fma3.c
new file mode 100644
index 000000000..cf34c6cb1
--- /dev/null
+++ b/numpy/distutils/checks/cpu_fma3.c
@@ -0,0 +1,8 @@
+#include <xmmintrin.h>
+#include <immintrin.h>
+
+int main(void)
+{
+    __m256 a = _mm256_fmadd_ps(_mm256_setzero_ps(), _mm256_setzero_ps(), _mm256_setzero_ps());
+    return (int)_mm_cvtss_f32(_mm256_castps256_ps128(a));
+}