[PATCH] D24593: Standford/Bubble sort code restructure

[Matthias Braun via llvm-commits]

MatzeB added a comment.

In https://reviews.llvm.org/D24593#543386, @evstupac wrote:

> The only public source I've found is:
>  http://classes.engineering.wustl.edu/cse465/docs/BCCExamples/stanford.c
>  But if it is the source, the benchmarks were already modified by submitter. Why we are unable to modify them now?
>
> Finally I see 2 solutions:
>
> 1. Benchmarks are owned by LLVM community and we can modify them.
> 2. Benchmarks are owned by someone else, we already modified them 14 years ago and do not want to get updates or make other modifications, because... initial modifications was approved by owner and there were no more updates. I can request one more approval from owner.


Yes the benchmark is probably a bad benchmark, has no official reference and it is unlikely to ever get updated. However given that we the very same will be true for a lot of source code in the wild, I don't see this as a bad thing. I still think it is more honest to live with the performance swings (we can explain them after all, so no need to blame the compiler or your patches) rather than modifying our benchmarks so the results look nicer.


https://reviews.llvm.org/D24593