This is the mail archive of the libc-alpha@sources.redhat.com mailing list for the glibc 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: gcc for glibc 2.3


On Tue, Aug 27, 2002 at 03:32:32PM -0700, Ulrich Drepper wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Roland McGrath wrote:
> > I moved __uselocale to GLIBC_2.3; all the __ names for the new locale fns have
> > always been prototyped.  Are you saying we should remove those prototypes
> > from the installed headers?
> 
> libstdc++ can and should indeed have own prototypes.  Thinking about it, 
> the __ variants of the new functions should be exported as 
> GLIBC_PRIVATE.  For packaging libstdc++ this would mean a bit more 
> effort (to avoid the errors caused by areference of a @GLIBC_PRIVATE 
> symbol).  But this is nothing new, it has to be done for glibc as well.

I tend to agree about the prototypes, but really don't think the symbols
should be GLIBC_PRIVATE. libstdc++ is not glibc and e.g. when libstdc++.a
is linked into an application yet libc.so is linked
dynamically, application writers would have to cope with GLIBC_PRIVATE too.
And as all but __uselocale are @@GLIBC_2.1 or something anyway, I think
__uselocale@@GLIBC_2.3 is good for consistency if nothing else.
If libc needs an private interface for it, it can come up with a different
name IMHO).

	Jakub


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