Analysis of Instruction Scheduling Strategies for VLIW Processors in order to Reduce Register Spilling