kquant_q8_1_rows

Function kquant_q8_1_rows 

Source
pub fn kquant_q8_1_rows(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    x_bf16: DevicePtr,
    out_q8: DevicePtr,
    m: u32,
    k: u32,
    stream: u64,
) -> Result<()>
Expand description

bf16 [m, k] -> block_q8_1 [m, k/32] (kquant_q8_1_rows_bf16). k % 32 == 0.