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: [PATCH] Additional features for -fms-extensions


Andrew Pinski <pinskia@physics.uc.edu> writes:

> The reason why this is an error and not a warning is because it will
> create "wrong" code and do what you expect in 3.4 and above.

> The reason why this error is here because the code will not work

> The reason why this is a warning and will abort at runtime is because it
> is undefined by the C++ standard and the ABI cannot handle when the 
> deconstructor is called.

These are not constructive responses.  The original poster said he was
aware that the C++ standard forbids all of these things.  He made a
legitimate feature request, with patches even; we may decide that we
don't want to add more bug-compatibility with Microsoft's compiler,
but that would be a policy decision, not a technical decision.

I have no opinion on the policy question.  I do care that people with
legitimate requests are not blown off with hostile responses.

zw


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