[cfe-dev] [clang-tidy] Issues with C++17 structured bindings

Tamir Zahavi-Brunner via cfe-dev cfe-dev at lists.llvm.org
Wed Jul 28 01:30:06 PDT 2021


The full contents of the file are:

#include <malloc.h>
#include <tuple>

int main()
{
    auto [ptr, num] = std::make_tuple(malloc(1), 1);
    free(ptr);

    return 0;
}

And the line I run which gives me the above errors is:

clang-tidy test.cpp -- --std=c++17

On Wed, Jul 28, 2021 at 11:06 AM Benics Bal√°zs <benicsbalazs at gmail.com>
wrote:

> Hi,
> It seems like its a Clang Static Analyzer checker warns about something.
>
> Could you please send the complete reproducer?
> I failed reproducing this.
>
> Regards,
> Balazs
>
>
> ------------------------------
> *From:* cfe-dev <cfe-dev-bounces at lists.llvm.org> on behalf of Tamir
> Zahavi-Brunner via cfe-dev <cfe-dev at lists.llvm.org>
> *Sent:* Tuesday, July 6, 2021 1:25:12 PM
> *To:* cfe-dev at lists.llvm.org <cfe-dev at lists.llvm.org>
> *Subject:* [cfe-dev] [clang-tidy] Issues with C++17 structured bindings
>
> Hello,
>
> Using clang-tidy-12, I'm getting false warnings when using structured
> bindings. There can actually be a variety of issues, but here is a very
> small example I've narrowed it down to:
>
> auto [ptr, num] = std::make_tuple(malloc(1), 1);
> free(ptr);
>
> For which I get the following warning:
>
> /tmp/test.cpp:7:5: warning: Potential leak of memory pointed to by
> '._M_head_impl' [clang-analyzer-unix.Malloc]
>     free(ptr);
>     ^
> /tmp/test.cpp:6:39: note: Memory is allocated
>     auto [ptr, num] = std::make_tuple(malloc(1), 1);
>                                       ^
> /tmp/test.cpp:7:5: note: Potential leak of memory pointed to by
> '._M_head_impl'
>     free(ptr);
>     ^
>
> Searching online, I saw that I'm not the first person to encounter such
> issues (
> https://stackoverflow.com/questions/62706928/clang-tidy-vs-structured-bindings)
> but I couldn't find any solution.
>
> Is this a bug in clang-tidy or am I missing something? Is there anything I
> can do to solve this other than stopping using structured bindings?
>
> Thanks,
> Tamir
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-dev/attachments/20210728/612184ea/attachment-0001.html>


More information about the cfe-dev mailing list