Optimize NEON shuffles #8
Description
They're too generic currently and inefficient. We can probably specialize most combinations using constructs such as
vcombine_f32(vget_high_f32(x), vget_low_f32(y))
vrev64q_f32(x)
etc.
{{ message }}
They're too generic currently and inefficient. We can probably specialize most combinations using constructs such as
vcombine_f32(vget_high_f32(x), vget_low_f32(y))
vrev64q_f32(x)
etc.