This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [Bug bootstrap/12276] [3.3/3.4 regression] Configure check forlibunwind does not work with cross
- From: Mark Mitchell <mark at codesourcery dot com>
- To: Andreas Schwab <schwab at suse dot de>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Fri, 03 Oct 2003 10:40:21 -0700
- Subject: Re: [Bug bootstrap/12276] [3.3/3.4 regression] Configure check forlibunwind does not work with cross
- Organization: CodeSourcery, LLC
- References: <20030914163803.12276.schwab@suse.de> <20031002221243.9285.qmail@sources.redhat.com> <jezngiz67y.fsf@sykes.suse.de>
On Fri, 2003-10-03 at 08:33, Andreas Schwab wrote:
> "mmitchel at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org> writes:
>
> > Apparently, this was considered to be desirable behavior; the ChangeLog says:
> >
> > 2002-10-02 David Mosberger-Tang <David.Mosberger@acm.org>
> >
> > * configure.in: Move sjlj-exceptions and --enable-libunwind-exceptions
> > before inclusion of config.gcc, but after configuring the compiler etc.
> > Determine default value for --enable-libunwind-exceptions based on
> > whether the host has a libunwind library (not guaranteed to be correct,
> > but it's a reasonable first guess and can always be overridden with an
> > explicit --enable/disable-libunwind-exceptions.
> >
> > Therefore, I've closed this bug. If you still think it's a bug, bring it up on
> > the GCC mailing list and see if you can get consensus about that.
>
> I don't think it is desirable to suddenly get a broken cross compiler just
> because someone installed libunwind on your host system.
>
> Tested both natively and with a ia64->m68k cross.
I agree that predictable behavior is good.
OK for mainline.
--
Mark Mitchell <mark@codesourcery.com>
CodeSourcery, LLC