================
@@ -2161,6 +2161,21 @@ static RValue EmitLoadOfMatrixLValue(LValue LV,
SourceLocation Loc,
return RValue::get(CGF.EmitLoadOfScalar(LV, Loc));
}
+RValue CodeGenFunction::EmitLoadOfAnyValue(LValue LV, AggValueSlot Slot,
+ SourceLocation Loc) {
+ QualType Ty = LV.getType();
+ switch (getEvaluationKind(Ty)) {
+ case TEK_Scalar:
+ return EmitLoadOfLValue(LV, Loc);
+ case TEK_Complex:
+ return RValue::getComplex(EmitLoadOfComplex(LV, Loc));
+ case TEK_Aggregate:
+ EmitAggFinalDestCopy(Ty, Slot, LV, EVK_RValue);
----------------
rjmccall wrote:
I believe the EVK argument here is mean to represent the kind of value that the
l-value represents. In that case, it's an l-value — we need to preserve the
underlying `va_list` storage and not treat it as something we own, because it's
possible to iterate varargs twice. I'm not sure it makes a practical
difference here because variadic argumets are required to be trivial, but it's
better to be accurate.
https://github.com/llvm/llvm-project/pull/94635
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
