This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Why does sparc build for you? or, why are make check results reported when it hasn't bootstrapped since May 5?
- From: arno at heho dot snv dot jussieu dot fr
- To: Jeff Sturm <jsturm at one-point dot com>
- Cc: ebotcazou at libertysurf dot fr, lucier at math dot purdue dot edu, gcc at gcc dot gnu dot org
- Date: 10 Jun 2003 19:35:07 +0200
- Subject: Re: Why does sparc build for you? or, why are make check results reported when it hasn't bootstrapped since May 5?
- References: <Pine.LNX.4.44.0306101248230.2871-100000@ops2.one-point.com>
Jeff Sturm <jsturm@one-point.com> writes:
> On Tue, 10 Jun 2003, Eric Botcazou wrote:
> > I meant: SPARC/Solaris would bootstrap more often if people commiting
> > destabilizing changes tested them on the platform. This is (almost) never the
> > case.
>
> Nice idea, difficult in practice. I've tested patches on Solaris 8
> that break on 7, or work with GNU as/ld but break on native tools, or vice
> versa. Then we still get bug reports for Solaris 2.6 or 2.5.1 users that
> I can't reproduce on my machines.
>
> For those and other reasons I find myself discouraged from testing on
> SPARC even though I have easy access to hardware. Getting regular
> contributors to test on SPARC doesn't seem too realistic.
>
> Such is probably the case for most non-GNU platforms.
for testing, I have fine experiences with 'half-GNU' :
I once (more or less) install gmake, gperf, bison, sed
[, gas, gld if available ]
and keep some /usr/out-of-the-way/old-gcc for bootstrapping
Solaris 2.5.1 bootstraps most of the time (I cvs update every
10 days and do a make install in /usr/out-of-the-way/new-gcc
whenever make check pleases me ...)
FYI, Arno