[PATCH] Fix a performance problem in gep(gep ...) merging

[Wei Mi]

> Well, Wei’s numbers seem to prove this is not an issue in practice.
> I would rather determine the right heuristic on actual motivating examples
> than sitting on the performance benefits Wei demonstrated.
>
> What would it take to convince you (not you in particular Chandler :)) that
> Wei’s heuristic is good enough for now?
>
> Would a diff of the binaries showing no regressions on the would test suite
> for, let say x86_64, arm, aarch64, and PPC, do?
> Note: I am talking about binary diff, no need to run.
>
> BTW, I assume Wei, that you already done something similar.
> What were the results?
>
> Cheers,
> -Quentin
>

Quentin, Thanks for the suggestion. I did some analysis about the code
size impact of the patch yesterday. Surprisingly I found the code size
of a lot of benchmarks with the patch increased by 0.1~0.3%. Then I
choosed a benchmark to look into and found the code size increase was
caused by more inlining -- some funcs have smaller code size with the
patch, so it is easier for those funcs to be inlined.

Wei.