Remarks on the Lefschetz standard conjecture for hyperkahler varieties