Struct Lanes16Word16 

#[repr(C, align(32))]
pub struct Lanes16Word16(pub [u16; 16]);

Sixteen lanes of Word16 (one 256-bit SIMD register) — the NTT coefficient lane vector. Carries [u16; 16]; the multiply-high is the SIGNED _mm256_mulhi_epi16 (the Montgomery butterfly’s mulhi). Ops have an AVX2 fast path proven byte-identical to the scalar lanes.

Tuple Fields

0: [u16; 16]

Implementations

impl Lanes16Word16

pub const LANES: usize = 16

The number of lanes.

pub const fn splat(x: u16) -> Self

Broadcast one value into all sixteen lanes.

pub fn from_words(s: &[Word16]) -> Self

Pack the first sixteen Word16s of a slice into a lane vector (shorter slices zero-fill).

pub fn to_words(self) -> [Word16; 16]

The lanes as sixteen Word16s.

pub fn lane(self, i: usize) -> Word16

Lane i (0-based) as a Word16.

pub fn add(self, o: Self) -> Self

Lane-wise wrapping add in ℤ/2¹⁶ (vpaddw).

pub fn sub(self, o: Self) -> Self

Lane-wise wrapping subtract in ℤ/2¹⁶ (vpsubw).

pub fn mullo(self, o: Self) -> Self

Lane-wise low 16 bits of the product (vpmullw) — interpretation-independent.

pub fn mulhi(self, o: Self) -> Self

Lane-wise high 16 bits of the SIGNED product (vpmulhw) — the Montgomery butterfly’s mulhi.

pub fn ntt_bcast_lo(self, h: usize) -> Self

Broadcast each 2h-block’s low h lanes into both of its halves — the within-vector NTT butterfly’s source-low duplication, at stride h ∈ {8,4,2}. h=8→vperm2i128(v,v,0x00); h=4→vpshufd(v,0x44); h=2→vpshufd(v,0xA0) (all per-128-bit-lane, so the multiple blocks packed in one register are handled at once).

pub fn ntt_bcast_hi(self, h: usize) -> Self

Broadcast each 2h-block’s high h lanes into both of its halves — the source-high duplication. h=8→vperm2i128(v,v,0x11); h=4→vpshufd(v,0xEE); h=2→vpshufd(v,0xF5).

pub fn ntt_blend(self, o: Self, h: usize) -> Self

Recombine the +/− halves of the within-vector butterfly: each 2h-block’s low h lanes from self, high h from o. h=8→vperm2i128(a,b,0x30); h=4→vpblendd(a,b,0xCC); h=2→vpblendd(a,b,0xAA).

Trait Implementations

impl Add for Lanes16Word16

type Output = Lanes16Word16

The resulting type after applying the + operator.

fn add(self, o: Self) -> Self

Performs the + operation.

impl Clone for Lanes16Word16

fn clone(&self) -> Lanes16Word16

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for Lanes16Word16

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Hash for Lanes16Word16

fn hash<__H: Hasher>(&self, state: &mut __H)

Feeds this value into the given Hasher.

fn hash_slice<H>(data: &[Self], state: &mut H)

where H: Hasher, Self: Sized,

Feeds a slice of this type into the given Hasher.

impl Mul for Lanes16Word16

type Output = Lanes16Word16

The resulting type after applying the * operator.

fn mul(self, o: Self) -> Self

Performs the * operation.

impl PartialEq for Lanes16Word16

fn eq(&self, other: &Lanes16Word16) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Sub for Lanes16Word16

type Output = Lanes16Word16

The resulting type after applying the - operator.

fn sub(self, o: Self) -> Self

Performs the - operation.

impl Copy for Lanes16Word16

impl Eq for Lanes16Word16

impl StructuralPartialEq for Lanes16Word16