[Bug c/61029] Bogus conversion warnings are issued on a sign change "a = -a;" operation

[manu at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=61029

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |manu at gcc dot gnu.org
         Resolution|INVALID                     |DUPLICATE

--- Comment #3 from Manuel López-Ibáñez <manu at gcc dot gnu.org> ---
(In reply to Dmitry Chichkov from comment #2)
> Results in:
>   -32768
>   -32768
>
> But still, it doesn't seem right that code like "a = -a;" has to be writted
> as a "a = static_cast<short int>(-a);" to avoid conversion warnings. That
> kills readability while not adding any value.

Also, the issue above is independent of the implicit conversion, it happens
also for int and INT_MIN. Unfortunately, there is some controversy among GCC
developers about whether this is the desired behavior of the warning or not. A
solution that may be accepted by all is to have a new warning option that does
not warn for cases where the target type is the same as the operand type. But
someone needs to implement this solution.

*** This bug has been marked as a duplicate of bug 40752 ***