rjmccall added inline comments.
================ Comment at: include/clang/AST/Decl.h:3631 + PassedIndirectly = true; + } + ---------------- I feel like this flag should be set by Sema for C++ types that have to be passed indirectly as well; it can then become the single point of truth for that information. ================ Comment at: lib/CodeGen/CGNonTrivialStruct.cpp:764 + Object = CGF->EmitObjCConsumeObject(QT, Object); + CGF->EmitARCStoreWeak(Addrs[DstIdx], Object, true); + } ---------------- I guess this is the most reasonable thing to do, given that we don't have an entrypoint for it. Please ask the ObjC runtime team to consider giving us one, though. We could pretty easily peephole assignments into `__weak` variables where the source is loaded from a `__weak` l-value / x-value, and Swift would probably be able to take advantage of it, too. You might want to go ahead and add `emitARCCopyAssignWeak` / `emitARCMoveAssignWeak` methods on CGF that do these operations and which can be optimized to use those entrypoints if/when they're added. ================ Comment at: lib/CodeGen/TargetInfo.cpp:1326 + if (RetTy.isPassedIndirectly() == QualType::APK_Struct) + return getNaturalAlignIndirect(RetTy); + ---------------- Can we combine this into a single function that covers both the C++ case and this new one? Otherwise it seems likely that individual targets over time will only add one or the other. Repository: rC Clang https://reviews.llvm.org/D44095 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
