Struct DecodeW8a8Plan 

pub struct DecodeW8a8Plan {
    pub rows: usize,
    pub n: u32,
    pub k: u32,
    pub ldc: u32,
    pub out_capacity_bytes: usize,
}

2026-09-25: The shape and output of one decode projection, as decode_w8a8_selected judges it. The route is taken only when all of these hold:

• family_armed: the caller’s super::CublasScope family (ssm or attn) is on.
• disabled is false (w8a8_decode_proj_disabled).
• rows is in DECODE_W8A8_ROWS.
• The weight is Fp8BlockScaled, since cuBLASLt is given its scales as a [N/128, K/128] grid.
• n and k are multiples of 128, and blk128x128_stride_ok(k) holds (k a multiple of 512), the weight-scale stride rule that metrale_gpu_runtime::cublaslt::scale_layout quotes.
• ldc >= n.
• The output holds the padded write extent (strided_out_extent_elems).
• The quantizer resolved, and the scratch holds m_pad rows in the active scale layout, adapter kernel and buffer included when it is K-major.

Fields

rows: usize

2026-09-25: Padded decode rows (the forward context’s n).

n: u32

2026-09-25: Output width of this projection.

k: u32

2026-09-25: Contract width of this projection.

ldc: u32

2026-09-25: Output row pitch in BF16 elements (n for a contiguous output).

out_capacity_bytes: usize

2026-09-25: Allocated size of the output buffer, in bytes.

Implementations

impl DecodeW8a8Plan

pub fn contiguous( rows: usize, n: u32, k: u32, out_capacity_bytes: usize, ) -> Self

2026-09-25: A contiguous [rows, n] output (GDN in_proj_qkvz and out_proj, attention o_proj).

pub fn strided( rows: usize, n: u32, k: u32, ldc: u32, out_capacity_bytes: usize, ) -> Self

2026-09-25: A strided output: rows ldc BF16 elements apart (attention Q/K/V into the [n, per_seq_qkv] multi-sequence QKV buffer).

pub fn m_pad(&self) -> u32

2026-09-25: The padded M cuBLASLt is handed.

pub fn write_extent_bytes(&self) -> usize

2026-09-25: Bytes of out this projection may write, pad rows included.

Trait Implementations

impl Clone for DecodeW8a8Plan

fn clone(&self) -> DecodeW8a8Plan

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for DecodeW8a8Plan

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

Formats the value using the given formatter.

impl Copy for DecodeW8a8Plan