This is the mail archive of the binutils@sources.redhat.com mailing list for the binutils project.


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

Re: [Proposed binutils PATCH] Re: Diagnosing an intricate C++ problem


> Date: Sun, 3 Sep 2000 13:28:42 +0200 (CEST)
> From: Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
> cc: rittle@rsch.comm.mot.com, gcc-patches@gcc.gnu.org,
>         binutils@sources.redhat.com
> 
> On 2 Sep 2000, Jason Merrill wrote:
> >> for which a simple `g++ x.cc` generates the following warnings:
> > Well, of course it does.  You're compiling it with different flags than
> > were used to build libstdc++, so the generated code is different.  That's
> > why this feature is pretty useless for code.
> 
> Yeah, but how can we diagnose the problem I raised in
>   http://gcc.gnu.org/ml/gcc/2000-07/msg00769.html
> ?
> 
> That's the main question, but the answer surely won't be easy! :-(

One way might be to modify the compiler so that the name of the
linkonce section is based on the token stream of the compiled
procedure.  That way, if two procedures have the same name but
different token streams, the linkonce sections will be different, and
you'll get a duplicate definition error.

-- 
- Geoffrey Keating <geoffk@cygnus.com>

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