AVX 256-bit equivalent for _mm_load1_ps

Question

With SSE, you can load one float from memory into all 4 __m128 slots with internal _mm_load1_ps ()

When using SIMD with 256-bit encryption with AVX, it seems there is no _mm256_load1_ps () to load one float from memory into all 8 vector slots.

Why is this an omission and what is the best way to do this?

Or even better: is there a way to load one float into the target slot 0..7 of the vector?

Answer 1

_ mm256_broadcast_ss is what you are looking for.