This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Mainline bootstrap failure in libjava, i686-pc-linux-gnu
- To: Geoff Keating <geoffk at geoffk dot org>
- Subject: Re: Mainline bootstrap failure in libjava, i686-pc-linux-gnu
- From: Phil Edwards <pedwards at disaster dot jaj dot com>
- Date: Mon, 9 Apr 2001 17:12:10 -0400
- Cc: gcc at gcc dot gnu dot org
- References: <20010406193001.A1583@codesourcery.com> <m2ae5t5l6l.fsf@kelso.bothner.com> <20010407144434.A19734@klomp.org> <20010409165731.E31897@disaster.jaj.com> <jm8zl994pk.fsf@geoffk.org>
On Mon, Apr 09, 2001 at 01:51:51PM -0700, Geoff Keating wrote:
> Phil Edwards <pedwards@disaster.jaj.com> writes:
>
> > I still don't understand why we pass -B$prefix/lib to the compiler during
> > the build, /before/ we install anything in $prefix. I've been told that
> > this is the Right Thing To Do, but I remain skeptical. :-(
>
> It's so that libraries that are installed in $prefix/lib are found.
Well, I knew /that/ much. :-)
> Otherwise,
> if you have a newlib install in $prefix/lib and you try to build the compiler
> by itself, it'll fail to build because it can't find the libraries.
>
> Now, that's one thing, but it should not be _preferring_ those
> libraries over the versions that we have just built...
AFAICT we always pass -Bvarious_build_dirs before -Bprefix/lib. So I don't
know how preinstalled libraries could be preferred over just-built ones,
unless we're installing them under different names and then using the
installation name during the link step instead of the build name.
(The -Bprefix thing used to bite me a lot in the past. It's seems to be
fixed now, but once bitten, eternally shy.)
Phil
--
pedwards at disaster dot jaj dot com | pme at sources dot redhat dot com
devphil at several other less interesting addresses in various dot domains
The gods do not protect fools. Fools are protected by more capable fools.