Function core_arch::x86::_mm_dp_pd

pub unsafe fn _mm_dp_pd(a: __m128d, b: __m128d, imm8: i32) -> __m128d

This is supported on x86 and target feature sse4.1 only.

Returns the dot product of two __m128d vectors.

imm8[1:0] is the broadcast mask, and imm8[5:4] is the condition mask. If a condition mask bit is zero, the corresponding multiplication is replaced by a value of 0.0. If a broadcast mask bit is one, the result of the dot product will be stored in the return value component. Otherwise if the broadcast mask bit is zero then the return component will be zero.

Intel's documentation