This is the mail archive of the
libc-alpha@sourceware.cygnus.com
mailing list for the glibc project.
Re: Build breakage
- To: Kaz Kylheku <kaz at ashi dot footprints dot net>
- Subject: Re: Build breakage
- From: Andreas Jaeger <aj at oss dot sgi dot com>
- Date: Fri, 21 Apr 2000 14:40:22 -0700
- Cc: libc-alpha at sourceware dot cygnus dot com
- References: <20000421140314.T9310@oss.sgi.com> <Pine.LNX.4.10.10004211414090.8790-100000@ashi.FootPrints.net>
On Fri, Apr 21, 2000 at 02:26:09PM -0700, Kaz Kylheku wrote:
> If you do a cvs diff after your update -A, are there any differences
> to the repository?
Not really:
$ cvs up -A
? ChangeLog.aj
P configure
M elf/Makefile
M manual/math.texi
>
> I think that maybe once a week, someone needs to do a clean ``cvs export'' and
> do a build from scratch, and then report breakages to the mailing list.
I'm running cvs update some times every day and rebuild glibc afterwards. Every
so often I start with a fresh build directory.
Ulrich is doing similiar testing - and I don't see directly why Flo and you have
problems but the two of us not.
Therefore I would expect glibc to build on ix86 without problems. Be assured
that whenever I see a problem on ix86, I'll fix it and send the patches or complain
about it.
AFAIK other problems (besides ix86 and MIPS which I take care of) do have problems.
Jakub just reported problems on alpha, and powerpc still has not changed the
dl-machine.h file and will therefore definitly fail.
> An exported code base much more closely matches what a user will eventually get
> in a release tarball.
The only difference are the CVS files - they shouldn't matter. The user gets something
different, because Ulrich runs make dist to create the tarball.
> I will try doing this for the next while to see how it goes.
Thanks, testers are always welcome.
Andreas
--
Andreas Jaeger
SuSE Labs aj@suse.de
private aj@arthur.rhein-neckar.de
currently: aj@oss.sgi.com