pub fn k4_ssd_state_passing(
intra_chunk_state_bnhpr: Tensor<5>,
da_chunk_end_bhn: Tensor<3>,
initial_state_bhpr: Tensor<4>,
) -> (Tensor<5>, Tensor<4>)Expand description
Based on the Kernel 4 Triton reference _state_passing_fwd_kernel (ssd_state_passing.py).
Returns:
- chunk_input_state_bnhpr
[used in K5][!]. - final_state_bhpr
[final output].