This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: libtool sending output to /dev/null [Was: Re: GCC build failed for native with your patch on 2002-06-02T22:30:03Z.]
- From: Marc Espie <espie at quatramaran dot ens dot fr>
- To: gcc at gcc dot gnu dot org
- Cc:
- Date: Tue, 4 Jun 2002 12:15:16 +0200
- Subject: Re: libtool sending output to /dev/null [Was: Re: GCC build failed for native with your patch on 2002-06-02T22:30:03Z.]
- Organization: Ecole Normale Superieure (quatramaran)
- References: <200206030144.g531iqx13347@maat.sfbay.redhat.com><20020603095755.GF27711@atrey.karlin.mff.cuni.cz><20020603061242.A9370@disaster.basement.lan><20020603101458.GG27711@atrey.karlin.mff.cuni.cz><20020603122106.GA2041@daikokuya.demon.co.uk><20020603090021.A10452@disaster.basement.lan><jmsn44uyyd.fsf - @desire.geoffk.org>
In article <or660zk5ly.fsf@free.redhat.lsd.ic.unicamp.br> you write:
>On Jun 3, 2002, Geoff Keating <geoffk@geoffk.org> wrote:
>
>> This is, by my count, the fourth time the regression tester has hit a
>> situation like this. Is there some way we can disable this in the GCC
>> tree?
>
>We could remove the redirection to /dev/null, but I'd rather have the
>output redirected to a file and only displayed in case of failure.
>Shouldn't be too hard to implement. In fact, this was discussed
>before, I gave pointers on how to do this a while ago, in the
>java@gcc.gnu.org mailing list, IIRC. It's been in my to-do list ever
>since.
How do you define `failure' ?
I'd much rather see the actual output of the commands libtool is running.
Some commands do issue useful diagnostics, which are not failures per-se,
but lead to tracking bugs in an easy way later.
In fact, OpenBSD ld warns about RSS relocations, which are coming to non-pic
code being linked as pic code (which leads to inefficient shared libraries).
If you redirect the warnings, you will never know things are wrong. You
will just notice that your shared libraries consume a fairly large amount
of memory.