This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: PR cpp/33907 Empty macro definitions not considered equal


On 29/11/2007, Tom Tromey <tromey@redhat.com> wrote:
> >>>>> "Manuel" == Manuel López-Ibáñez <lopezibanez@gmail.com> writes:
>
> Manuel> This patch fixes PR33907. The conclusion is that warning about A being
> Manuel> redefined for this testcase is annoying:
> Manuel> #define A(a)
> Manuel> #define A(b)
>
> Is this a change from 4.2?

The only change is that pedantic warnings (pedwarns) are errors by
default in the C++ CPP (as they are in the C++ front-end). Everything
is explained in detail in the PR.

> It seems to me that if we can be pedantic without unduly affecting
> existing code, then we ought to be.

So then close the bug as invalid. Perhaps Martin Michlmayr could
comment on this. See
http://www.cyrius.com/journal/2007/05/11#gcc-4.3-pedwarn

But I don't really see how the redefinition in this case can make any
difference at all. And if you really want strict ISO conformance and
thus strict ISO diagnostics, even the pointless annoying ones, that is
what -pedantic is for, and you will get the error in that case.

You are the maintainer, take a decision. In this case, I would vote
for my patch since I wasted the time discussing, submitting and
testing it but it is your call and I don't care so much either way.

Cheers,

Manuel.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]