martong added a comment. I am attaching the performance measurement results:
F20834663: removemember_revert_with_another_fixpoint.png <https://reviews.llvm.org/F20834663> F20834673: stats.html <https://reviews.llvm.org/F20834673> There is no any noticeable difference. I am going to commit this as is because there is a push to revert D114619 <https://reviews.llvm.org/D114619> (because of a flaky test) and this commit does that. On the other hand this is not a pure revert, so I apologize for the reviewers. If we were to purely revert D114619 <https://reviews.llvm.org/D114619> then we'd have to revert its dependent child patch as well. I have a strong confidence in this patch since I had a verbal discussion about this with @steakhal and he verified the basic idea. I am going to address my comment about `simplifyUntilFixpoint` in a follow-up patch. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D114887/new/ https://reviews.llvm.org/D114887 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
