aaron.ballman added inline comments.
================ Comment at: clang/lib/Analysis/ThreadSafety.cpp:2091-2097 + if (auto *CE = dyn_cast<CastExpr>(E)) + if (CE->getCastKind() == CK_NoOp) + E = CE->getSubExpr()->IgnoreParens(); + if (auto *CE = dyn_cast<CastExpr>(E)) + if (CE->getCastKind() == CK_ConstructorConversion || + CE->getCastKind() == CK_UserDefinedConversion) + E = CE->getSubExpr(); ---------------- aaronpuchert wrote: > aaron.ballman wrote: > > I almost wonder if we should just turn this into a while loop rather than > > two casts in a specific order. However, I can't think of a situation where > > we'd need the loop, so I think this is fine. At the very least, it's > > incremental progress! > They should always be in that order, or at least until we get to the > `CXXConstructExpr` or `CXXMemberCallExpr` that we want (there could be > another standard conversion sequence for the argument inside of that, but > that's not our concern here). The reason is that this constructor or > conversion function call is the `CK_ConstructorConversion` or > `CK_UserDefinedConversion`, respectively, while the remaining standard > conversion sequence (to which the `NoOp` belongs) [happens after/above > that](https://eel.is/c++draft/over.best.ics#over.ics.user). Ah, thank you for confirming my intuition! Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D129752/new/ https://reviews.llvm.org/D129752 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
