[Interest] Destructor appearing twice in the stack trace?

Etienne Sandré-Chardonnal etienne.sandre at m4x.org
Tue Nov 10 08:05:08 CET 2015


By the way, I would have expected QQbject to "deparent" C before the
destroyed() signal is sent to connected slots. This would have prevented
the crash there.

Etienne

2015-11-10 8:01 GMT+01:00 Etienne Sandré-Chardonnal <etienne.sandre at m4x.org>
:

> Hi,
>
> I have fixed the crash problem, but the double destructor is still there.
> For the record, what is the compiler mechanism behind this?
>
> The crash issue was simple : I have three QObjects with parents A -> B ->
> C (grandchild)
> The C destructor was sending destroyed() signal  to B slot, which was
> sending a signal to A's slot, which was deleting B, still a parent of C.
> Replacing "delete B" by "B->deleteLater()" fixed it. But the double
> destructor in stack is unrelated and still there.
>
> Thanks
>
>
>
>
> 2015-11-09 18:19 GMT+01:00 Thiago Macieira <thiago.macieira at intel.com>:
>
>> On Monday 09 November 2015 16:21:44 Etienne Sandré-Chardonnal wrote:
>> > Dear all,
>> >
>> > I am debugging an app which crashes when an object is being deleted. The
>> > debuggers halt on a segmentation fault inside QObject destructor. The
>> > object deletion was requested by a call to deleteLater()
>> >
>> > Here is what I get in the stack trace:
>> >
>> > 0    QObject::~QObject    qobject.cpp    993    0x7ffff62c7580
>> > 1    RendererInterface::~RendererInterface    rendererinterface.cpp
>> > 119    0x60664f
>> > 2    RendererInterface::~RendererInterface    rendererinterface.cpp
>> > 137    0x60667e
>> > 3    qDeleteInEventHandler    qobject.cpp    4391    0x7ffff62d06e5
>> >
>> > How is it possible to have the destructor twice in the stack?? At the
>> > beginning I was thinking this was a GDB bug (GDB/MinGW under windows is
>> > buggy) but I get the same result when compiling the app under linux.
>> >
>> > Any idea how this can happen?
>>
>> It depends on what code the compiler generated. It's entirely possible,
>> but I
>> can't also rule out a problem.
>>
>> --
>> Thiago Macieira - thiago.macieira (AT) intel.com
>>   Software Architect - Intel Open Source Technology Center
>>
>> _______________________________________________
>> Interest mailing list
>> Interest at qt-project.org
>> http://lists.qt-project.org/mailman/listinfo/interest
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.qt-project.org/pipermail/interest/attachments/20151110/56e061d3/attachment.html>


More information about the Interest mailing list