This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: GCC target deprecation
- From: Joe Buck <jbuck at synopsys dot com>
- To: Rick Sustek <rsustek at cisco dot com>
- Cc: gcc at gcc dot gnu dot org, zack at codesourcery dot com
- Date: Mon, 30 Dec 2002 12:39:31 -0800
- Subject: Re: GCC target deprecation
- References: <3E074DDA.F4F96AEF@cisco.com> <20021223152713.A3289@synopsys.com> <3E08AF08.5EEFBC0A@cisco.com>
On Mon, Dec 23, 2002 at 09:54:34AM -0800, Rick Sustek wrote:
> > > Still using:
> > > m68k-sun-sunos*
> > > on Sun3 hardware platform.
I wrote:
> > Interesting. Are you using current or near-current GCC releases?
> > We haven't seen any build success reports or test results for this
> > platform, and it would be nice to have some.
> >
> > I assume you are cross-compiling?
On Tue, Dec 24, 2002 at 11:01:28AM -0800, Rick Sustek wrote:
> I'm on gcc 2.95.3.
> Cross compiling from two hosts: sparc-sun-solaris2.6 and some intel
> Redhat linux host.
>
> Planning on moving into the gcc 3.* world, but was having frustrating
> time bootstrapping
> my way up the release train, just for native compiling on the
> sparc-sun-solaris2.6 host,
> and could not get any of the 3.* releases to build. Thus, that is on
> hold for me now.
[ I'm back now after the holidays ]
For 3.2.1, we have successful reports for sparc-sun-solaris2.6, both
with binutils and with native as and ld. There were some issues with
earlier 3.x releases.
> But, it is good (???) to know that I might expect build trouble for a
> cross compiler to
> the m68k-sun-sunos* target, should I be able to get to that point. ;=)
Generally speaking, anything that no one has tested for a while tends to
break, especially for older OSes that have peculiarities that must be
worked around (e.g. SunOS). The good news, however, is that the breakage
often has a trivial fix. So don't lose hope if you get a failure, instead
please ask for help, showing the exact failure that you get. Also,
cross-compilation tends to be a bit tricky, and some things are different
now; better to ask for help than to suffer alone.
> BTW, consider my level of interest in the m68k-sun-sunos* target to be
> relatively low,
> i.e., I have no crucial business case hinging on its continued support
> in gcc. My interest
> is higher in the more generic m68k-* targets: m68k-elf and m68k-aout,
> which are not
> on your current list of deprecation targets.
Joel Sherill reported successful builds for m68k-elf and m68k-coff for
gcc 3.2.1 + binutils 2.13.1; see
http://gcc.gnu.org/ml/gcc/2002-12/msg00326.html
He didn't say anything about m68k-aout, perhaps because he had no
interest.