[PATCH] D115497: [Inline] Disable deferred inlining

[Nikita Popov via Phabricator via llvm-commits]

nikic added a comment.

In D115497#3192305 <https://reviews.llvm.org/D115497#3192305>, @mtrofin wrote:

> LGTM, and a question to reviewers: how should we proceed with understanding if we should actually disable/remove deferral (because, judging just on my datapoints, it's basically a no-op)

My plan would be to land this, and if no major regressions are reported follow up after the LLVM 14 release to drop the option and implementation. If major regressions are reported, it would be good to add reduced test cases for them, as this functionality is currently untested (apart from one incidental CGSCC invalidation test).


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D115497/new/

https://reviews.llvm.org/D115497