libtool sending output to /dev/null [Was: Re: GCC build failed for native with your patch on 2002-06-02T22:30:03Z.]

Geoff Keating geoffk@geoffk.org
Mon Jun 3 09:08:00 GMT 2002


Phil Edwards <phil@jaj.com> writes:

> On Mon, Jun 03, 2002 at 01:21:06PM +0100, Neil Booth wrote:
> > This is how it died for me.  Why do we pipe stderr to /dev/null?
> 
> Look at the previous [two] command[s].  The makefile is calling libtool
> to build a .lo file.  Libtool, in turn, runs (and echos) the build twice.
> The first time, it passes -fPIC -DPIC and lets the user see diagnostics.
> The second time, it doesn't pass any PIC options, and sinks the output on the
> assumption that the user won't want to see the (presumably) same diags twice.
> 
> The assumption is a reasonably good one, but it falls down when the
> presence/absence of -fPIC makes a difference.  Which seems to be the case
> here, since there's no other difference.

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?

-- 
- Geoffrey Keating <geoffk@geoffk.org> <geoffk@redhat.com>



More information about the Gcc-regression mailing list