[Bug c++/109494] New: inline const variables interfere with source_location

[Bug c++/109494] New: inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

            Bug ID: 109494
           Summary: inline const variables interfere with source_location
           Product: gcc
           Version: 12.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: oliver.rosten at googlemail dot com
  Target Milestone: ---

The bug, (for which I'm not sure if it is a front end problem or a library
problem), is reproduced here: https://github.com/ojrosten/SourceLoc

The project comprises three files in the Test directory:

Main.cpp
Test.cpp
Test.hpp

Both the cpps depend on the hpp. The latter contains an unused variable

inline const std::string foo{};

The presence of this seems to cause:

1. The appearance of a warning: "ld: warning: direct access in function ...
from file ... to global weak symbol ... from file ... means the weak symbol
cannot be overridden at runtime..."

2. std::source_location::current() to misbehave: building and running causes
the path to Main.cpp to be output twice, whereas it should just be printed
once, with the path to Test.cpp appearing second.

Any of the following cures both of the problems:

1. Removing foo;

2. Removing inline;

3. Replacing const with constexpr

In the much more complex project where I first encountered this, I reliably got
a segmentation fault. I was able to cure this by removing inline in a handful
of places.

[Bug target/109494] inline const variables interfere with source_location

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2023-04-13
     Ever confirmed|0                           |1
          Component|c++                         |target
           Keywords|diagnostic                  |wrong-code
             Status|UNCONFIRMED                 |WAITING

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Can you provide the information as requested on https://gcc.gnu.org/bugs/ ?
At least the output of "gcc -v".

Also can you attach the testcase and not use cmake as cmake makes it hard to
figure out the exact command line which is being invoked, a shell script or a
simple make file should be enough?

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #2 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
Created attachment 54851
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54851&action=edit
Preprocessed file

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #3 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
Created attachment 54852
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54852&action=edit
Preprocessed file for Test.cpp

[Bug target/109494] inline const variables interfere with source_location

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

Patrick Palka <ppalka at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |ppalka at gcc dot gnu.org

--- Comment #4 from Patrick Palka <ppalka at gcc dot gnu.org> ---
I can't reproduce the linker warning or bad output with GCC 12.2 or trunk on
x86_64-pc-linux-gnu:

$ g++ -std=c++20 -fext-numeric-literals Main.ii Test.ii
$ ./a.out
/Users/Claire/Dropbox/SourceLoc/Test/Main.cpp
/Users/Claire/Dropbox/SourceLoc/Test/Test.cpp

[Bug target/109494] inline const variables interfere with source_location

[ppalka at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #5 from Patrick Palka <ppalka at gcc dot gnu.org> ---
(In reply to Patrick Palka from comment #4)
> I can't reproduce the linker warning or bad output with GCC 12.2 or trunk on
> x86_64-pc-linux-gnu:
> 
> $ g++ -std=c++20 -fext-numeric-literals Main.ii Test.ii
> $ ./a.out
> /Users/Claire/Dropbox/SourceLoc/Test/Main.cpp
> /Users/Claire/Dropbox/SourceLoc/Test/Test.cpp
Oops, it was already established this is a darwin-specific issue, sorry for the
noise.

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #6 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
I've added a simple makefile to the project:
https://github.com/ojrosten/SourceLoc

[Bug target/109494] inline const variables interfere with source_location

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #7 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
You still didn't provide the full output.

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #8 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
Created attachment 54870
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=54870&action=edit
Output from gcc -v

[Bug target/109494] inline const variables interfere with source_location

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #9 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
You most likely should report this to Homebrew too:

https://github.com/Homebrew/homebrew-core/issues

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #10 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
Can do, but I don't have sufficient understanding to understand why! Can you
kindly explain?

[Bug target/109494] inline const variables interfere with source_location

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Oliver Rosten from comment #10)
> Can do, but I don't have sufficient understanding to understand why! Can you
> kindly explain?

Because that is where you got the GCC you are using from. It is listed in the
configure line:
"--with-bugurl=https://github.com/Homebrew/homebrew-core/issues"

[Bug target/109494] inline const variables interfere with source_location

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|WAITING                     |UNCONFIRMED
     Ever confirmed|1                           |0

[Bug target/109494] inline const variables interfere with source_location

[oliver.rosten at googlemail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494

--- Comment #12 from Oliver Rosten <oliver.rosten at googlemail dot com> ---
Yes, I am aware that I homebrewed gcc. What I don't understand is the
extent to which this a homebrew problem and not a gcc problem. Forgive my
ignorance but I would like to understand what submitting this issue to
homebrew implies/achieves. I simply don't have much awareness of the
broader context.

On Sat, 15 Apr 2023 at 22:51, pinskia at gcc dot gnu.org <
gcc-bugzilla@gcc.gnu.org> wrote:

> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109494
>
> --- Comment #11 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
> (In reply to Oliver Rosten from comment #10)
> > Can do, but I don't have sufficient understanding to understand why! Can
> you
> > kindly explain?
>
> Because that is where you got the GCC you are using from. It is listed in
> the
> configure line:
> "--with-bugurl=https://github.com/Homebrew/homebrew-core/issues"
>
> --
> You are receiving this mail because:
> You reported the bug.