Modify the backend, or do a binary translation from one to the other and test. If `lea` is the predominate instruction, there might be microcode optimizations that favor `lea` over `movl`. My hunch is that is will be mostly the same barring overflowing the instruction cache. The microps should compile to the same instruction stream.
How would one find this out? Sounds like it would be fun to figure out how to develop all possible reasonably compact instruction combinations to achieve the same basic block and then compare timings.
lea also doesn't modify the flags register. At one point this meant there was a wider choice of execution units it could be scheduled on than arithmetic instructions that required a full ALU.
