pub fn w4a16_gemv_dual(
gpu: &dyn GpuBackend,
kernel: KernelHandle,
input: DevicePtr,
weight1: &QuantizedWeight,
output1: DevicePtr,
weight2: &QuantizedWeight,
output2: DevicePtr,
n: u32,
k: u32,
stream: u64,
) -> Result<()>Expand description
W4A16 dual GEMV: two projections sharing the same BF16 input, one launch.
blockIdx.z selects projection 0 vs 1. Both N dimensions must be equal.
Grid: (ceil(N/4), 1, 2) Block: (256, 1, 1)