Merge pull request #142 from mkinsner/additional_loop_control_bits

Allocate three loop control bits for an upcoming Intel extension