[Bug c/92286] Possible improvement for -Wduplicated-cond warning

[Bug c/92286] Possible improvement for -Wduplicated-cond warning

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92286

David Binderman <dcb314 at hotmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dcb314 at hotmail dot com

--- Comment #4 from David Binderman <dcb314 at hotmail dot com> ---
Is there a simpler case which is worth addressing, where a condition
is a simple opposite of a predecessor ?

Something like:

void f2( int a, int b)
{
        if (a < b)
                g( a);
        else if (a >= b)
                g( b);
        else
                g( a + b);
}

Static analyser cppcheck can find this:

feb23b.cc:20:13: style: Expression is always true because 'else if' condition
is
 opposite to previous condition at line 18. [multiCondition]
 else if (a >= b)
            ^
feb23b.cc:18:8: note: first condition
 if (a < b)
       ^
feb23b.cc:20:13: note: else if condition is opposite to first condition
 else if (a >= b)
            ^

[Bug c/92286] Possible improvement for -Wduplicated-cond warning

[dcb314 at hotmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92286

--- Comment #5 from David Binderman <dcb314 at hotmail dot com> ---
A quick grep suggests two problems in gcc source code:

trunk.git/gcc/ada/sysdep.c:423:26: style: Expression is always true because
'else if' condition is opposite to previous condition at line 415.
[multiCondition]
trunk.git/libsanitizer/sanitizer_common/sanitizer_allocator_primary64.h:547:27:
style: Expression is always true because 'else if' condition is opposite to
previous condition at line 538. [multiCondition]

and 34 cases in the linux-6.0 kernel.

[Bug c/92286] Possible improvement for -Wduplicated-cond warning

[marxin at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92286

--- Comment #6 from Martin Liška <marxin at gcc dot gnu.org> ---
> trunk.git/libsanitizer/sanitizer_common/sanitizer_allocator_primary64.h:547:
> 27: style: Expression is always true because 'else if' condition is opposite
> to previous condition at line 538. [multiCondition]

This one is correct, but the code assumes there's no fallback edge:

...
    } else if (chunk_size > page_size) {
      // One chunk covers multiple pages, Some chunks are crossing page
      // boundaries. Some pages contain one chunk, some contain two.
      full_pages_chunk_count_max = 2;
      same_chunk_count_per_page = false;
    } else {
      UNREACHABLE("All chunk_size/page_size ratios must be handled.");
    }

Thus I'm going to leave it.