pub fn eval_systolic_os( os: &SystolicOs, a: &[i64], w: &[i64], bits: u32, ) -> Vec<i64>
Reference int8 GEMM for the output-stationary tile: O[i][j] = Σ_t A[i][t]·W[t][j] (wrapping i64).
O[i][j] = Σ_t A[i][t]·W[t][j]