[Bug c/50486] New: No warning at signed -> unsigned casting

[Bug c/50486] New: No warning at signed -> unsigned casting

[kirill at shutemov dot name]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50486

             Bug #: 50486
           Summary: No warning at signed -> unsigned casting
    Classification: Unclassified
           Product: gcc
           Version: 4.6.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: kirill@shutemov.name


$ cat 1.c
enum e {
        A,
        B,
        C
};

int a(enum e e)
{
        if (e < 0)
                return 1;
        return 0;
}

int b(void)
{
        return a(-1);
}

Since enum e has only positive values GCC states that enum is unsigned and
eliminates the if-check. I can see warning about it with -Wtype-limits.
But GCC generates no warning at a(-1), so it silently casts -1 to unsigned.

[Bug c/50486] No warning at signed -> unsigned casting

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50486

Manuel López-Ibáñez <manu at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011-09-22
                 CC|                            |manu at gcc dot gnu.org
     Ever Confirmed|0                           |1

--- Comment #1 from Manuel López-Ibáñez <manu at gcc dot gnu.org> 2011-09-22 15:02:46 UTC ---
-1 is converted to 4294967295 very early in the FE, but I don't know why or
where.

I think Wsign-conversion could warn about this. I will confirm it then, but I
am not working on it.

[Bug c/50486] No warning at signed -> unsigned casting

[egallager at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=50486

Eric Gallager <egallager at gcc dot gnu.org> changed:

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

--- Comment #2 from Eric Gallager <egallager at gcc dot gnu.org> ---
(In reply to Manuel López-Ibáñez from comment #1)
> -1 is converted to 4294967295 very early in the FE, but I don't know why or
> where.
> 
> I think Wsign-conversion could warn about this. I will confirm it then, but
> I am not working on it.

...or, now that there's a -Wenum-conversion flag, how about that?

[Bug c/50486] No warning at signed -> unsigned casting

[manu at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=50486

--- Comment #3 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
Clang warns:

 <source>:16:18: warning: implicit conversion changes signedness: 'int' to
'enum e' [-Wsign-conversion]
        return a(-1);
               ~ ^~

Not sure why gcc doesn't but it should.

[Bug c/50486] Missed -Wsign-conversion with signed -> unsigned casting and enums

[egallager at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=50486

Eric Gallager <egallager at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|No warning at signed ->     |Missed -Wsign-conversion
                   |unsigned casting            |with signed -> unsigned
                   |                            |casting and enums

--- Comment #4 from Eric Gallager <egallager at gcc dot gnu.org> ---
updating the title a bit