transpose_block_scale

Function transpose_block_scale 

Source
pub fn transpose_block_scale(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    src: DevicePtr,
    dst: DevicePtr,
    n_blocks: u32,
    k_blocks: u32,
    stream: u64,
) -> Result<()>
Expand description

2026-09-25: Transpose FP32 block scales: [N/128, K/128] to [K/128, N/128].