A bug in GCC that may cause memory leaks in valid C++ programs

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by MereInterest on Thursday April 27, 2017 @06:21PM (#14215285)

I don't think it seems like a code smell at all. For example, I could be adding together multiple vectors, some of which are created as temporaries.

    auto total_offset = Vector{x_offset, y_offset} + global_offset;

Granted, in this case, there are no resources being held by such a Vector class, but having immediately-destroyed temporary objects is not necessarily a code smell.

[ Parent | on HN ]

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by wfunction on Thursday April 27, 2017 @06:26PM (#14215325)

That's definitely not "immediately destroyed", there's operator+ being called between construction and destruction.

I didn't realize this can trigger the bug though, so if it does, I should clarify I didn't mean to include it in my comment. I was only referring to cases where there is no intervening operation between construction and destruction.

[ Parent | on HN ]

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by SolarNet on Thursday April 27, 2017 @06:45PM (#14215446)

It's a minimal testcase; minimal. I thought the blog post was pretty clear that it had nothing to do with how the object was used after (e.g. if it was immediately destructed or not) and even linked to a real world example that would disprove that.

Hence why your holier than thou attitude about not writing smelly code is just asinine and not helpful. It plays into a mindset that people who write "good code" don't have to worry about 'esoteric' bug reports (e.g. like security advisories).

[ Parent | on HN ]

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by jmgao on Friday April 28, 2017 @03:00AM (#14217411)

There's probably far more reasonable cases of temporaries being immediately destructed than you'd expect.

One that springs to mind is gtest. It defines a bunch of macros like ASSERT_TRUE that return a temporary object that has operator<< overloads to let you add additional information to the assertion failure. For example:

    ASSERT_NE(-1, open("/foo", O_RDONLY)) << "failed to open file: " << strerror(errno);

It's perfectly legitimate to just go ASSERT_TRUE(1 != 2), though.

[ Parent | on HN ]

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by wfunction on Friday April 28, 2017 @03:08AM (#14217455)

Like I said:

> That's definitely not "immediately destroyed", there's operator<< being called between construction and destruction.

[ Parent | on HN ]

Re:A bug in GCC that may cause memory leaks in valid C++ pro... by yorwba on Friday April 28, 2017 @07:39AM (#14218589)

The point is that you have the option of using operator<<, but if you don't exercise that option, the temporary object is destroyed immediately.

[ Parent | on HN ]