Module simd_wide 

True vector SIMD using the wide crate for f64x2 operations

This module provides vectorized implementations of mathematical operations using the wide crate which provides portable SIMD types that compile on stable Rust.

Strategy: Process f64 values in pairs using f64x2 vectors. 8 f64 values = 4 f64x2 vectors. This gives actual hardware vectorization while maintaining portability.

Functions

simd_abs_8_wide

Vectorized absolute value using f64x2

simd_acos_8_wide

Vectorized inverse cosine using f64x2

simd_add_8_wide

Vectorized element-wise addition using f64x2

simd_asin_8_wide

Vectorized inverse sine using f64x2

simd_atan2_8_wide

Vectorized inverse tangent (atan2) using f64x2

simd_clamp_8_wide

Vectorized clamp operation using f64x2

simd_cos_8_wide

Vectorized cosine for 8 f64 values using f64x2

simd_madd_8_wide

Vectorized fused multiply-add using f64x2

simd_mul_8_wide

Vectorized element-wise multiplication using f64x2

simd_recip_8_wide

Vectorized reciprocal using f64x2

simd_sin_8_wide

Vectorized sine for 8 f64 values using f64x2

simd_sin_cos_8_wide

Vectorized sin and cos simultaneously using f64x2

simd_sqrt_8_wide

Vectorized square root using f64x2