gdn_carry_wy

Function gdn_carry_wy 

Source
pub fn gdn_carry_wy(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    h_table: DevicePtr,
    query: DevicePtr,
    key: DevicePtr,
    value: DevicePtr,
    gate: DevicePtr,
    beta: DevicePtr,
    output: DevicePtr,
    carry_base: DevicePtr,
    slot_tab: DevicePtr,
    pend: DevicePtr,
    seq_floats: u32,
    batch_size: u32,
    num_k_heads: u32,
    num_v_heads: u32,
    qk_stride: u32,
    v_stride: u32,
    gb_stride: u32,
    k_dim: u32,
    engaged_flag: DevicePtr,
    stream: u64,
) -> Result<()>
Expand description

2026-09-26: Carried-state verify (gdn_carry_wy{2,3,4} or its _lazy form, chosen by the caller’s kernel). It applies each sequence’s pending rows (pend[slot_tab[b]]) to its copy of H and writes H back (the lazy form only when this verify’s rows might not fit behind them in the stash, pending + K > GDN_CARRY_CAP), writes the output, and stashes the rows it verified at carry_base + slot * seq_floats. Engaged word b becomes 2 after a write-back, else 1.