[Bug c/112826] New: gcc rejects the valid code and reports undefined reference

[Bug c/112826] New: gcc rejects the valid code and reports undefined reference

[jwzeng at nuaa dot edu.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112826

            Bug ID: 112826
           Summary: gcc rejects the valid code and reports undefined
                    reference
           Product: gcc
           Version: 14.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: jwzeng at nuaa dot edu.cn
  Target Milestone: ---

Hi, the following code seems valid. Gcc rejects it while clang accepts.

Compiler explorer: https://godbolt.org/z/sPK9sYqc6

```c
$ cat test.c
int* a();
static int b() { int *c = a();}
int main() {}
$
$ gcc test.c
/tmp/cchtt1Nc.o:test.c:function b: error: undefined reference to 'a'
collect2: error: ld returned 1 exit status
$
$ clang test.c
test.c:2:31: warning: non-void function does not return a value [-Wreturn-type]
    2 | static int b() { int *c = a();}
      |                               ^
1 warning generated.
$
$ gcc --version
gcc (GCC) 13.1.0
Copyright (C) 2023 Free Software Foundation, Inc.
$
$ clang --version
clang version 16.0.0
Target: x86_64-unknown-linux-gnu
Thread model: posix
```

[Bug c/112826] gcc rejects the valid code and reports undefined reference

[arsen at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112826

Arsen Arsenović <arsen at gcc dot gnu.org> changed:

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

--- Comment #1 from Arsen Arsenović <arsen at gcc dot gnu.org> ---
I can't assert the validity of this code, but -O1 does elide the function as
expected.

[Bug c/112826] gcc rejects the valid code and reports undefined reference

[xry111 at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112826

Xi Ruoyao <xry111 at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |INVALID
             Status|UNCONFIRMED                 |RESOLVED
                 CC|                            |xry111 at gcc dot gnu.org

--- Comment #2 from Xi Ruoyao <xry111 at gcc dot gnu.org> ---
N1570 6.9 p2:

"If an identifier declared with external linkage is used in an expression
(other than as part of the operand of a sizeof or _Alignof operator whose
result is an integer constant), somewhere in the entire program there shall be
exactly one external definition for the identifier"

and annex J.2 explicitly lists "An identifier with external linkage is used,
but in the program there does not exist exactly one external definition for the
identifier" as undefined behavior.

Note that you cannot interpret "used" as "really executed at runtime" because
it's impossible to solve halting problem, so such an "interpretation" will
render the entire standard nonsense.

[Bug c/112826] gcc rejects the valid code and reports undefined reference

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112826

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=14105

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
Note PR 14105 is the opposite request but both are invalid as both behavior
with and without optimization is allowed by the C standard.

[Bug c/112826] gcc rejects the valid code and reports undefined reference

[jwzeng at nuaa dot edu.cn]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=112826

--- Comment #4 from Junwei Zeng <jwzeng at nuaa dot edu.cn> ---
Thank you for your explanation.