[Bug c++/101601] New: [12 Regression] -Wall is reported as a build error (not warning): error: arrays of functions are not meaningful

[Bug c++/101601] New: [12 Regression] -Wall is reported as a build error (not warning): error: arrays of functions are not meaningful

[slyfox at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

            Bug ID: 101601
           Summary: [12 Regression] -Wall is reported as a build error
                    (not warning): error: arrays of functions are not
                    meaningful
           Product: gcc
           Version: 12.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: slyfox at gcc dot gnu.org
  Target Milestone: ---

The example is extracted from wine-6.13 build failure against gcc-master.

Here is the smaller example:

$ cat a.c
    // x86_64-pc-linux-gnu-gcc -m32 -c -o a.o a.c -Wall -pipe -O2
    typedef void f(void);
    void args(void*);
    f *SSCall_fun;
    void SSCall(void) { args(&SSCall_fun + 1); }

$ ./xgcc -B.  -m32 -c -o a.o a.c -Wall -pipe -O2
a.c: In function 'SSCall':
a.c:5:6: error: arrays of functions are not meaningful
    5 | void SSCall(void) { args(&SSCall_fun + 1); }
      |      ^~~~~~

Note1: -Wall turns successful build into build failure.

I think it's not expected: it should still be a warning (or an unconditional
error).

$ ./xgcc -B.  -m32 -c -o a.o a.c -Wall -pipe
$ ./xgcc -B.  -m32 -c -o a.o a.c -pipe -O2
$ gcc-11.1.0  -m32 -c -o a.o a.c -Wall -pipe -O2

Note2: -O2 vs. -O0 causes warning to go away (maybe it's ok).

$ ./xgcc -B. -v
Reading specs from ./specs
COLLECT_GCC=./xgcc
COLLECT_LTO_WRAPPER=./lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: /home/slyfox/dev/git/gcc/configure --build=x86_64-pc-linux-gnu
--host=x86_64-pc-linux-gnu --target=x86_64-pc-linux-gnu
--enable-languages=c,c++ --disable-bootstrap --with-multilib-list=m64
--prefix=/tmp/gcc-build/../gcc-native-quick-installed --disable-nls
--without-isl --disable-libsanitizer --disable-libvtv --disable-libgomp
--disable-libstdcxx-pch --disable-libunwind-exceptions CFLAGS='-O1 '
CXXFLAGS='-O1 '
Thread model: posix
Supported LTO compression algorithms: zlib zstd
gcc version 12.0.0 20210723 (experimental) (GCC)

[Bug middle-end/101601] [12 Regression] -Wall is reported as a build error (not warning): error: arrays of functions are not meaningful

[mpolacek at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

Marek Polacek <mpolacek at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c++                         |middle-end
   Last reconfirmed|                            |2021-07-23
     Ever confirmed|0                           |1
                 CC|                            |mpolacek at gcc dot gnu.org
             Status|UNCONFIRMED                 |NEW

--- Comment #1 from Marek Polacek <mpolacek at gcc dot gnu.org> ---
This comes from array_bounds_checker::check_array_bounds -> not a FE issue.

[Bug middle-end/101601] [12 Regression] -Wall is reported as a build error (not warning): error: arrays of functions are not meaningful

[slyfox at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

Sergei Trofimovich <slyfox at gcc dot gnu.org> changed:

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

--- Comment #2 from Sergei Trofimovich <slyfox at gcc dot gnu.org> ---
Bisected down to r12-2132 "Correct handling of variable offset minus constant
in -Warray-bounds [PR100137]"

[Bug middle-end/101601] [12 Regression] -Warray-bounds triggers error: arrays of functions are not meaningful

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |msebor at gcc dot gnu.org
             Status|NEW                         |ASSIGNED
           Keywords|                            |diagnostic
   Target Milestone|---                         |12.0
            Summary|[12 Regression] -Wall is    |[12 Regression]
                   |reported as a build error   |-Warray-bounds triggers
                   |(not warning): error:       |error: arrays of functions
                   |arrays of functions are not |are not meaningful
                   |meaningful                  |

--- Comment #3 from Martin Sebor <msebor at gcc dot gnu.org> ---
-Warray-bounds triggers the error while formatting the type of SSCall_fun.  For
simplicity, it converts the type of each singleton object into a one-element
array  of its type, but the code is wrong for function pointers.

[Bug middle-end/101601] [12 Regression] -Warray-bounds triggers error: arrays of functions are not meaningful

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

--- Comment #4 from CVS Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Martin Sebor <msebor@gcc.gnu.org>:

https://gcc.gnu.org/g:b9cbf8c9e0bc72f59b643165247fae646560aadd

commit r12-2582-gb9cbf8c9e0bc72f59b643165247fae646560aadd
Author: Martin Sebor <msebor@redhat.com>
Date:   Wed Jul 28 16:14:38 2021 -0600

    Correct -Warray-bounds handling if function pointers [PR101601].

    Resolves:
    PR middle-end/101601 - -Warray-bounds triggers error: arrays of functions
are not meaningful

            PR middle-end/101601

    gcc/ChangeLog:

            * gimple-array-bounds.cc (array_bounds_checker::check_mem_ref):
Remove
            a pointless test.
            Handle pointers to functions.

    gcc/testsuite/ChangeLog:

            * g++.dg/warn/Warray-bounds-25.C: New test.
            * gcc.dg/Warray-bounds-85.c: New test.

[Bug middle-end/101601] [12 Regression] -Warray-bounds triggers error: arrays of functions are not meaningful

[msebor at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101601

Martin Sebor <msebor at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #5 from Martin Sebor <msebor at gcc dot gnu.org> ---
Fixed in r12-2582.