This is the mail archive of the libstdc++@gcc.gnu.org mailing list for the libstdc++ 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: verbose terminate() on by default


On Tue, Dec 24, 2002 at 01:59:06AM +0100, Gabriel Dos Reis wrote:
> Phil Edwards <phil@jaj.com> writes:
> | Actually, I'd rather just make a libsupc++/vterminate.cc, rather than
> | making eh_terminate.cc pull in the I/O library.  Right now eh_terminate.cc
> | is very minimal.  The verbose handler and all the extra baggage will be
> | pulled in only if eh_terminate.cc still initializes __terminate_handler to
> | the verbose function -- and that's something the user could easily change.
> | Or perhaps we'll provide some way of doing it.
> 
> Sorry, I don't get that.  
> 
> Since the default behaviour is to be verbose, the I/O part will be
> pulled in, whether user changes it to something different of not.
> Unless the linker is very smart, but then in that speculative case, it
> would also have enough information to notice that
> __verbose_terminate_handler() is not necessary (since static and not
> used in the current module) and to eliminate it altogether with the
> I/O bits. 

I didn't explain myself properly.  I'm not relying on the linker to be smart.
What I want is to make this scenario work:

    http://gcc.gnu.org/ml/libstdc++/2002-12/msg00308.html


Phil

-- 
I would therefore like to posit that computing's central challenge, viz. "How
not to make a mess of it," has /not/ been met.
                                                 - Edsger Dijkstra, 1930-2002


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