This is the mail archive of the
libc-alpha@sources.redhat.com
mailing list for the glibc project.
Re: __errno_location export
- From: Momchil Velikov <velco at fadata dot bg>
- To: Jakub Jelinek <jakub at redhat dot com>
- Cc: Roland McGrath <roland at redhat dot com>, libc-alpha at sources dot redhat dot com
- Date: 02 Dec 2002 00:39:15 +0200
- Subject: Re: __errno_location export
- References: <874ra6v8e4.fsf@merlin.maxx.bg><200211250000.gAP00OE06993@magilla.sf.frob.com><20021125011005.B7036@sunsite.ms.mff.cuni.cz><871y51trpi.fsf@merlin.maxx.bg><20021201222652.V1310@sunsite.ms.mff.cuni.cz>
>>>>> "Jakub" == Jakub Jelinek <jakub@redhat.com> writes:
Jakub> On Sun, Dec 01, 2002 at 10:22:17PM +0200, Momchil Velikov
Jakub> wrote:
>> >>>>> "Jakub" == Jakub Jelinek <jakub@redhat.com> writes:
>>
Jakub> 2002-11-25 Jakub Jelinek <jakub@redhat.com>
>>
Jakub> * include/errno.h (__set_errno): Define as errno = val
Jakub> unconditionally.
>> This way I get undefined references to errno, most probably
>> because of the following in sysdeps/generic/errno.c
Jakub> The above mentioned change has nothing to do with export of
Jakub> errno.
The above change replaces references to ``__errno_location'' with
references to ``erno''. It has much to do with whether ``errno'' is
exported or not.
Jakub> It depends if your port is using __errno_location in
Jakub> bits/errno.h as definition of errno, or not. If yes, then
Jakub> not exporting errno as link time symbol is desirable,
Jakub> otherwise you IMHO shouldn't be using
Jakub> sysdeps/generic/errno.c.
Well then, I guess I should #define errno (*__errno_location ()) in
the port's bits/errno.h.
Thanks,
-velco