transpose_fp8

Function transpose_fp8 

Source
pub fn transpose_fp8(
    gpu: &dyn GpuBackend,
    kernel: KernelHandle,
    src: DevicePtr,
    dst: DevicePtr,
    n: u32,
    k: u32,
    stream: u64,
) -> Result<()>
Expand description

2026-09-25: Transpose an FP8 weight on the GPU: B[N, K] to B_t[K, N].