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: [C++ PATCH] Allow some function pointer conversions


Jason Merrill wrote:

On 20 Oct 2004 12:25:12 -0500, Gabriel Dos Reis <gdr@cs.tamu.edu> wrote:



Mark Mitchell <mark@codesourcery.com> writes:

| Nathan Sidwell wrote:
| | I think that if we want to implement DR195, I'd be happiest if we just
| removed the diagnostic. (Before my patch, things were inconsistent;
| we issued a pedwarn for reinterpret_cast, but not for a C-style cast.
| I made the two match up.) I think that we could just remove the
| diangostic altogether to implement DR195.
| | We could also turn it into a warning, which might be the most
| conservative choice; we would still conform to TC1, but also implement
| DR195.


If we believe we really want to implement DR195, then I would
recommend we do not remove the diagnostic, but make it a warning.



My preference would be for a warning only under -pedantic. This pattern is supported on all of our platforms, and is in wide use. Warning about it would just be useless noise unless people are specifically interested in extreme portability. A hard error is unacceptable.

Good; I think we're in agreement.

Nathan, I see that your patch for DR195 has gone in. Shall I revert that and replace the current pedwarn with a warning, thereby implementing the above solution? Or, do you want to do that? Or, do you not like this plan at all?

Thanks,

--
Mark Mitchell
CodeSourcery, LLC
(916) 791-8304
mark@codesourcery.com


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