[llvm] r285106 - [AArch64] Adjust the cost model for Exynos M1.

[Renato Golin via llvm-commits]

On 25 October 2016 at 22:03, Evandro Menezes <e.menezes at samsung.com> wrote:
> Short of going into the details of the indirect branch predictor in M1, the
> change that improved the heuristics creating jump tables
> (https://reviews.llvm.org/D25212) results in a tad more conditional
> branches, which tend to perform better than indirect branches on most
> processors.  This heuristic change called for a revision of the limit of the
> size of jump tables for M1, based on a suite of benchmarks, such as SPEC
> CPU2000 and CPU2006.

That would have been a much better commit message. :)

Maybe just adding something like "SPEC2k +x% SPEC2k6 +y%" would have
been nice, too.

cheers,
--renato