[PATCH] D101844: [MicroBenchmarks] Add initial loop vectorization benchmarks.

[Michael Kruse via Phabricator via llvm-commits]

Meinersbur added a comment.

In D101844#2736635 <https://reviews.llvm.org/D101844#2736635>, @jroelofs wrote:

> My understanding was that googlebench only measures timing in the `for (auto _ : state)` loop, and that the setup before / after isn't counted.

Correct; I misread the call structure thinking that `run_fn_autovec` was doing the comparison.


Repository:
  rT test-suite

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D101844/new/

https://reviews.llvm.org/D101844