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]

Re: Restore __eprintf


On 31-May-2001, Alexandre Oliva <aoliva@redhat.com> wrote:
> On May 31, 2001, Fergus Henderson <fjh@cs.mu.oz.au> wrote:
> 
> > You may also need to remove any object files or static libraries
> > compiled with older versions of gcc, since they might contain
> > references to __eprintf().
> 
> This shouldn't be a problem in general, assuming you're going to use
> these object files and static libraries to create executables.
> Indeed, if you use them to create shared libraries, they won't find
> __eprintf() in libgcc_s, but for executables, __eprintf() will be
> present in the static libgcc.

Earlier in this thread, IIRC, people said that (a) on Irix, `-static' was
effectively ignored and (b) when linking with shared libgcc, the static
libgcc.a wasn't searched.  From these statements, I concluded (perhaps
wrongly -- if so please correct me) that __eprintf() would not be
found in the static libgcc.a when linking executables on IRIX,
since the static libgcc.a wouldn't be searched.

> > Or alternatively, how about putting it in the shared glibc for all
> > architectures, marking it as deprecated, and then removing it in GCC 4.0?
> 
> There's no reason to introduce this binary compatibility problem right
> now.

I don't understand.  I'm talking about NOT introducing a binary compatibility
problem now, and instead delaying its introduction until later, e.g. GCC 4.0,
by which time it won't be a problem anymore, since GCC 3.* won't generate
references to __eprintf like GCC 2.* did.

Or are you saying that removing __eprintf in GCC 4.0 would cause a problem,
but that removing it in GCC 3.0 won't?  If so, I don't understand why.

> Perhaps we should even install libeprintf.a, containing
> only eprintf.o, so that the work-around would be even simpler:
> -leprintf.

That would help, but the need for any work-around is going to be a bit of a
pain.

-- 
Fergus Henderson <fjh@cs.mu.oz.au>  |  "I have always known that the pursuit
                                    |  of excellence is a lethal habit"
WWW: <http://www.cs.mu.oz.au/~fjh>  |     -- the last words of T. S. Garp.


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