================
@@ -592,10 +590,14 @@ void AggExprEmitter::EmitArrayInit(Address DestPtr,
llvm::ArrayType *AType,
// observed to be unnecessary.
if (endOfInit.isValid()) Builder.CreateStore(element, endOfInit);
}
-
- LValue elementLV = CGF.MakeAddrLValue(
- Address(element, llvmElementType, elementAlign), elementType);
+ Address address = Address(element, llvmElementType, elementAlign);
+ LValue elementLV = CGF.MakeAddrLValue(address, elementType);
EmitInitializationToLValue(Args[i], elementLV);
+ // Schedule to emit element cleanup if we see a branch in the array
+ // initialisation expression.
+ if (CGF.needsBranchCleanup(dtorKind))
+ CGF.pushDestroy(BranchInExprCleanup, address, elementType,
----------------
efriedma-quic wrote:
Are you sure you generated the IR correctly? Note that clang marks everything
optnone at -O0 by default.
That said, I guess -O0 codegen is a potential issue; I don't think we run any
passes that would clean up the dead stores in that case. I'd only be concerned
about the case where we don't actually emit the cleanup; trying to minimize the
codesize of the cleanup code isn't important at -O0.
To address the case of no cleanups, maybe we could generate the stores lazily?
Basically, save where the stores should be, but don't generate them until we
know the cleanup is actually used. Or equivalently, emit the stores, then
erase them if the cleanup isn't emitted.
https://github.com/llvm/llvm-project/pull/80698
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
